bert-finetuned-phishing-ONNX

ONNX export of ealvaradob/bert-finetuned-phishing, packaged for the Lemonade router classifier backend (ort-server).

• Base model: ealvaradob/bert-finetuned-phishing
• Task: text-classification
• Labels: benign, phishing

Files

file	purpose
model.onnx	the exported model (input_ids/attention_mask → logits)
tokenizer.json	the original HuggingFace tokenizer
manifest.json	task / labels / normalization for ort-server
export.py	the exact script used to produce & validate these files

Validation after export

Exported from source with 🤗 Optimum and validated against the original PyTorch model on fixtures (ONNX Runtime CPU vs HF): max softmax delta 0.0 (0 = identical).

CPU-EP latency (ONNX Runtime, single input): ~26.38 ms p50 @ 12 tokens.

Reproduce

pip install "optimum[onnxruntime]" transformers torch onnxruntime sentencepiece
python export.py ealvaradob/bert-finetuned-phishing ./out --task text-classification

See validation.json for the recorded parity result.

License

Follows the base model ealvaradob/bert-finetuned-phishing; refer to it for terms.